Pith. sign in

DiffusionNOCS: Managing Symmetry and Uncertainty in Sim2Real Multi-Modal Category-level Pose Estimation

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

This paper addresses the challenging problem of category-level pose estimation. Current state-of-the-art methods for this task face challenges when dealing with symmetric objects and when attempting to generalize to new environments solely through synthetic data training. In this work, we address these challenges by proposing a probabilistic model that relies on diffusion to estimate dense canonical maps crucial for recovering partial object shapes as well as establishing correspondences essential for pose estimation. Furthermore, we introduce critical components to enhance performance by leveraging the strength of the diffusion models with multi-modal input representations. We demonstrate the effectiveness of our method by testing it on a range of real datasets. Despite being trained solely on our generated synthetic data, our approach achieves state-of-the-art performance and unprecedented generalization qualities, outperforming baselines, even those specifically trained on the target domain.

fields

cs.CV 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

One2Any: One-Reference 6D Pose Estimation for Any Object

cs.CV · 2025-05-07 · conditional · novelty 6.0

One2Any achieves state-of-the-art single-reference 6D pose accuracy on real benchmarks by conditioning a U-Net decoder on a learned reference embedding and a reference-frame coordinate map.

citing papers explorer

Showing 1 of 1 citing paper.

  • One2Any: One-Reference 6D Pose Estimation for Any Object cs.CV · 2025-05-07 · conditional · none · ref 24 · internal anchor

    One2Any achieves state-of-the-art single-reference 6D pose accuracy on real benchmarks by conditioning a U-Net decoder on a learned reference embedding and a reference-frame coordinate map.